TRANSFUSION COMPLICATIONS A prion reduction filter does not completely remove endogenous prion infectivity from sheep blood  lez,2 Sandra McCutcheon,1 A. Richard Alejo Blanco,1 Boon Chin Tan,1 Lorenzo Gonza Stuart Martin,2 Gary Mallinson,3 Nigel E. Appleford,3 Marc L. Turner,4 Jean C. Manson,1 and E. Fiona Houston1

BACKGROUND: Variant Creutzfeldt-Jakob disease (vCJD) is a transmissible spongiform encephalopathy affecting humans, acquired initially through infection with bovine spongiform encephalopathy (BSE). A small number of vCJD cases have been acquired through the transfusion of blood from asymptomatic donors who subsequently developed vCJD. Filter devices that selectively bind the infectious agent associated with prion disease have been developed for removal of infection from blood. This study independently assessed one such filter, the P-CAPT filter, for efficacy in removing infectivity associated with the BSE agent in sheep blood. The sheep BSE model has previously been used to evaluate the distribution of infectivity in clinically relevant blood components. This is the first study to assess the ability of the P-CAPT filter to remove endogenous infectivity associated with blood components prepared from a large animal model. STUDY DESIGN AND METHODS: Paired units of leukoreduced red blood cells (LR-RBCs) were prepared from donors at the clinical stage of infection and confirmed as having BSE. One cohort of recipients was transfused with LR-RBCs alone, whereas a parallel cohort received LR and P-CAPT–filtered RBCs (LRRBCs-P-CAPT). RESULTS: Of 14 recipients, two have been confirmed as having BSE. These sheep had received LR-RBCs and LR-RBCs-P-CAPT from the same donor. CONCLUSIONS: The results indicate that, after leukoreduction and P-CAPT filtration, there can still be sufficient residual infectivity in sheep RBCs to transmit infection when transfused into a susceptible recipient.

T

he identification of variant Creutzfeldt-Jakob disease (vCJD) in humans in 19961,2 is believed to be linked to the consumption of meat products from cattle infected with bovine spongiform encephalopathy (BSE).3 The United Kingdom has shown the highest incidence of vCJD in the world with 177 definite or probable cases reported up until November 3, 2014.4,5 During the early stages of the vCJD outbreak, the risk of disease transmission via iatrogenic routes such as blood transfusion, organ donation, and surgical procedures was unknown. However, data from multiple prioninfected models were beginning to show that blood collected before the onset of clinical signs could transmit

ABBREVIATIONS: BSE 5 bovine spongiform encephalopathy; DMNV 5 dorsal motor nuclei of the vagus nerve; LR 5 leukoreduced; LR-RBC-P-CAPT 5 leukoreduced and P-CAPT–filtered RBCs; PK 5 proteinase K; qPCR 5 quantitative real-time polymerase chain reaction; RAMALT 5 rectoanal mucosa-associated lymphoid tissues; vCJD 5 variant Creutzfeldt-Jakob disease. From the 1Neurobiology Division, The Roslin Institute, University of Edinburgh, and 2Animal and Plant Health Agency, Lasswade Laboratory, Edinburgh, UK; 3Bristol Institute for Transfusion Sciences, Bristol, UK; and the 4University of Edinburgh and SNBTS, Edinburgh, UK. Address reprint requests to: E. Fiona Houston, Neurobiology Division, the Roslin Institute, Royal (Dick) School of Veterinary Studies, University of Edinburgh, Easter Bush, Edinburgh EH25 9RG, UK; e-mail: [email protected]. Supported by UK Blood Services (Contract Reference NBS1024/G/AP). SM and ARAB contributed equally to this work. Received for publication January 13, 2015; revision received March 10, 2015; and accepted March 13, 2015. doi:10.1111/trf.13145 C 2015 AABB V

TRANSFUSION 2015;55;2123–2133 Volume 55, September 2015 TRANSFUSION 2123

MCCUTCHEON ET AL.

infection following experimental inoculation of laboratory hosts.6-13 This was confirmed using large animal models (sheep and deer), which have provided convincing evidence that blood transfusion is also an efficient route of prion transmission.14-20 The Transfusion Medicine Epidemiology Review was set up in 2006 to investigate whether there was evidence that CJD or vCJD may have been transmitted via the blood supply.21 This review has identified three cases of vCJD21-26 in UK residents who had been transfused with nonleukoreduced red blood cells (RBCs) from donors who were later confirmed to have died from vCJD. In common with all confirmed clinical cases of vCJD to date, these individuals were homozygous for methionine (M) at Codon 129 of the PRNP gene (129MM). A fourth recipient of nonleukoreduced RBCs, who died of causes unrelated to vCJD, but tested positive for diseaseassociated PrP (PrPd) and infectivity in the spleen, was methionine/valine (M/V) heterozygous at Codon 129 (129MV).27 More recently, a 129MV patient with hemophilia who had received Factor VIII prepared from plasma pools known to contain donations from a vCJD donor also tested positive for PrPd in spleen samples, but again had no history of neurologic disease.28 The latter two cases raise the possibility that individuals with 129MV and 129VV genotypes are also susceptible to infection with vCJD, but may have longer incubation periods or be asymptomatic carriers of infection. This is supported by data from infection of mice transgenic for the human PRNP gene.29 Moreover, recent large-scale surveys of human appendices identified samples from individuals of 129MM, 129MV, and 129VV genotypes positive for PrPd by immunohistochemistry, resulting in prevalence estimates for subclinical vCJD infection of up to 1 in 2000 of the UK population.30 In the absence of a validated diagnostic test to screen for vCJD infection, these asymptomatic individuals present a potential risk for further human-to-human transmission via iatrogenic routes. To reduce the likelihood of vCJD transmission via blood products, a number of protective measures have been adopted in the United Kingdom, including donor deferral, importation of plasma, and leukoreduction (filtration to remove white blood cells [WBC]) of all components used in human transfusion medicine.31,32 Continued follow-up in the Transfusion Medicine Epidemiology Review study has not yet identified any evidence of infection in patients who received leukoreduced (LR) blood components from known vCJD cases. However, several experimental models have shown that not all endogenous infectivity in blood is removed after leukoreduction.17,32,33 Therefore, additional risk reduction measures have been considered for the targeted removal of the infectious agent (PrPd) from human blood components before transfusion or preparation of plasma-derived 2124 TRANSFUSION Volume 55, September 2015

therapeutic products.34-37 One novel approach is prion filtration, which aims to remove abnormal PrP from blood by selective binding to affinity ligands or resins, using devices such as the MacoPharma P-CAPT filter,32,38,39 the Pall Corporation Leukotrap filter,40-42 and the Asahi KASEI combination filter.41 Many of the studies evaluating the efficacy of these prion reduction filters have used brain spiked into blood or blood components38-40,42 as a way to assess the extent of prion removal. However, this approach may not replicate the effects of filtration on endogenous infectivity in blood. The aim of this study was to assess the efficacy of a specific prion removal filter, the CE-marked P-CAPT filter, in removing endogenous infectivity associated with the BSE agent from sheep blood. The effects of the filter have previously been tested using both spiked39 and endogenous infectivity32 associated with the 263 K hamster scrapie strain. Studies to assess the effect of processing human LR-RBCs, using the P-CAPT filter, identified no issues surrounding the clinical safety of P-CAPT–filtered products43 and found no evidence of adverse effects with respect to component storage, stability, viability, or other measures of RBC quality.44-47 We have applied the P-CAPT filter to LR units of RBCs prepared from sheep infected with BSE and compared the qualitative effects, in terms of disease transmission and detection of abnormal prion protein, with the transfusion of paired units of LR-RBCs alone. In this sheep model, the distribution of both infectivity and of abnormal prion protein in peripheral lymphoid tissues is similar to that of humans affected with vCJD.48,49 The relevance of this model in assessing the transfusion-related transmission of prion diseases in sheep has been documented in several publications.14-17 We have previously shown the presence of infectivity in both RBCs and LR equivalents, when blood was collected from sheep at a preclinical stage of infection. In this study, we report on infectivity associated with LR-RBCs that have been filtered using the P-CAPT filter.

MATERIALS AND METHODS Experimental animals The animal experiments were approved by The Institute for Animal Health and The Roslin Institute’s (University of Edinburgh) Animal Welfare and Ethical Review Committees and experiments conducted according to the regulations of the UK Home Office Animals (Scientific Procedures) Act 1986. The sheep were sourced from the DEFRA scrapie-free flock50 and were ARQ/ARQ at Codons 136, 154, and 171 of the sheep PRNP gene (genotypes determined as previously described15). The genotypes at Codon 141 are indicated in Table 1, whereby the effects relate to survival

EVALUATION OF P-CAPT FILTER

TABLE 1. Details of donor and recipient sheep Donor details* Sheep ID

PRNP Codon 141 genotype

Survival period (days postinfection)†

N220

FF

797

N218

FF

840

N245

LF

959

N180

FF

817

N258

LL

944

N178

LF

1168

N259

LF

1076

Recipient details Sheep ID

PRNP Codon 141 genotype

Survival period (days postinfection)†

LR-RBCs

P524

LF

732

LR-RBCs-P-CAPT LR-RBCs LR-RBCs-P-CAPT LR-RBCs LR-RBCs-P-CAPT LR-RBCs LR-RBCs-P-CAPT LR-RBCs LR-RBCs-P-CAPT LR-RBCs LR-RBCs-P-CAPT LR-RBCs LR-RBCs-P-CAPT

P511 P548 P466 P467 P461 Q391 Q397 P494 P304 P541 P465 P473 P280

FF FF LF LF LF LF LF LF LF LF FF LF LF

1049 1130 1984‡ 573 1981‡ 1970‡ 1970‡ 1927‡ 1927‡ 1774‡ 1773‡ 1710‡ 1710‡

Component

Reason for cull

Status

Welfare (suspect clinical) Clinical Welfare NA Welfare NA NA NA NA NA NA NA NA NA

BSE 1 BSE 1 BSE – Alive BSE – Alive Alive Alive Alive Alive Alive Alive Alive Alive

* All donors were culled, upon reaching defined clinical endpoints associated with BSE infection, immediately after 2 units of whole blood was collected (i.e., on the day of blood donation). † Survival periods were calculated as the number of days between the time of infection (by oral route for donors or by transfusion for recipients) and euthanasia or October 3, 2014, for sheep that are still alive (indicated by the ‡).

times associated with experimental scrapie and BSE infection.51,52

Donor sheep and blood collection For this study, seven sheep were used as blood donors. This cohort was part of a larger group of sheep used as blood donors in an ongoing blood transfusion study (referred to as the leukoreduction study) and were maintained as previously described.17 Donors were orally challenged with 5 g of brain homogenate prepared from BSEinfected cattle (our reference BBB4/5, APHA reference SE1909/BBP2) and were clinically monitored until they reached defined humane clinical endpoints associated with BSE infection, as previously described.15 Immediately before euthanasia, 2 units of whole blood (1 unit 5 450 mL 6 10% [vol/vol]) were collected from each donor into quadruple, bottom-and-top blood packs (Fresenius Hemocare, NPBI, Bad Homburg, Germany) fitted with inline leukoreduction filters, containing 63 mL of CPD-A1.

Preparation of LR-RBCs and LR and P-CAPT– filtered RBCs The methods of preparing blood components, including P-CAPT–filtered equivalents for transfusion were written in conjunction with staff at the Scottish National Blood Transfusion Service (SNBTS) to mimic procedures used for the processing of blood from patients and to achieve compliance with instructions provided for use of the prion reduction filter. An illustrative figure showing the blood collection and component preparation process is shown

in Fig. S1 (available as supporting information in the online version of this paper). Immediately after donation (Day 0), the 2 whole blood units were processed to RBCs, as previously described.17 RBCs were LR (using the inline T3953 filter) and were stored overnight at 48C (without shaking). The following day (Day 1), both LR-RBC units were warmed to 228C and mixed gently using a seesaw shaker for 30 to 60 minutes to ensure homogeneity. The 2 LRRBC units were then pooled into an oversized blood bag with tubing connected to two separate transfer bags (referred to as the Christmas tree configuration pack). This approach was employed to make each of the final products to be transfused as homogenous as possible not only in terms of their cellular content but also in terms of the distribution of infectivity associated with the BSE agent in blood. After pooling and gentle mixing to ensure homogeneity, the pooled LR-RBCs were allowed to flow into the transfer packs, resulting in 2 separate units of LRRBCs of approximately equal volume. One of the 2 LRRBC units was transfused to a recipient sheep without further processing, after cross-matching of donor and recipient blood samples, as previously described.15 The other LR-RBC unit was processed using the PCAPT filter according to the manufacturer’s instructions. A P-CAPT filter and associated transfer pack was docked (using sterile conditions) to the LR-RBC unit, with tubing separating the two (of a standard length of 15 cm). A stainless-steel adjustable clamp was fitted to the tubing immediately below the P-CAPT filter and closed completely before filtration. This was used to manage the flow Volume 55, September 2015 TRANSFUSION 2125

MCCUTCHEON ET AL.

rate of blood during the filtration process. The snap valve fitted inside the tubing between the LR-RBC bag and the P-CAPT filter was opened to allow the flow of LR-RBCs into the P-CAPT filter. Next, the adjustable stainless-steel clamp was opened by a few (4-5) millimeters allowing controlled flow of LR-RBCs through the P-CAPT filter and into the associated transfer pack, with the aim of ensuring that the filtration time took between 40 and 60 minutes (in line with recommendations from the manufacturer relating to blood contact time with the P-CAPT filter). The resultant leukoreduced and P-CAPT–filtered RBCs (LRRBCs-P-CAPT) component was transfused to a recipient sheep, after cross-matching of donor and recipient blood samples.

Specifications of LR-RBC and LR-RBC-P-CAPT components used for transfusion The volumes of each unit of whole blood collected, LRRBC and LR-RBC-P-CAPT components prepared (mL) were determined by subtracting the weight of an empty collection bag from the filled weight of each component and dividing by the specific gravities of 1.06 for whole blood and RBCs, as previously described.17 Similarly, the methods for determining WBC counts and the distribution of plasma in the RBCs, LR-RBCs, and LR-RBC-PCAPT components have already been described in detail.17 The time (in minutes) taken for leukoreduction and P-CAPT filtration was also recorded.

WBC evaluation using quantitative real-time PCR of sheep CD45 CD45 antigen is a universal and specific marker of WBCs in blood. In addition to the BD Leukocount technique previously described,17 residual numbers of WBCs from sheep blood samples were determined by quantitative real-time polymerase chain reaction (qPCR) of CD45 genomic DNA (gDNA) by two separate methods using a sequence detection system (Applied Biosystems 7000, Life Technologies, Grand Island, NY). Small-volume aliquots (0.5-1 mL) of each unit of RBCs (before leukoreduction), LR-RBCs, and LR-RBCs-P-CAPT were stored at 2208C at the time of preparation. gDNA was extracted from 50 mL of sheep blood using the sheep blood kit (Invitrogen ChargeSwitch gDNA, Life Technologies), and two PCR methods were used to quantitate gDNA, Power SYBR Green PCR (Applied Biosystems, Life Technologies, Paisley, UK) and Taqman Universal PCR (Life Technologies), according to the manufacturer’s protocols. The primers were designed using the mRNA sequence of sheep CD45, which was identified from a blast search of the Sheep Genome v2.0 (http://www.livestockgenomics.csiro.au/sheep/oar2. 0.php) against the bovine homolog of human CD45 (NM_001206523.1). The forward primer was 50 GCACCTACATCGGCATCGA and the reverse primer 50 2126 TRANSFUSION Volume 55, September 2015

CGTAGACGTCCACCTTGTTCTCT. The number of WBCs in each sample was calculated from the amplified gDNA by comparison against a standard curve of qPCR amplified CD45 from samples containing gDNA of known concentration (0.6 pg-60 ng) assuming that one cell contained 6.6 pg of DNA.

Recipient sheep As for donor sheep, recipient sheep were maintained and monitored for clinical signs of BSE infection until they reached defined humane endpoints or were culled for other reasons. Samples of rectoanal mucosa-associated lymphoid tissues (RAMALT) were collected at intervals to monitor for preclinical infection, as previously described.53 At postmortem, samples of brain and lymphoid tissues (tonsil, spleen, ileal Peyer’s patch, mesenteric lymph node, prescapular lymph node) were collected and either fixed in neutral buffered formalin or frozen at 2808C.

Detection of disease-associated PrP in tissues by Western blotting and immunohistochemistry Proteinase K (PK)-resistant PrPSc in brain and lymphoid tissues was detected using established Western blot methods, probing with the monoclonal antibody (MoAb) ROSBC6 (0.5 mg/mL) as previously described.17,54,55 Autoradiography film (Hyperfilm, GE Healthcare, Buckinghamshire, UK) was scanned and the digital image processed using a computer program (Adobe Photoshop, Adobe Systems, Inc., San Jose, CA). Image processing was restricted to cropping, preparation of composite figures, and annotation. Formalin-fixed sections of brain and lymphoid tissues were also analyzed for PrPd deposition by immunohistochemistry using MoAbs ROS-IH955 and BG4, as previously described. The MoAbs were applied and incubated overnight—ROS-IH9 (0.5 mg/mL at 48C) or BG4 (1.25 mg/ mL at ambient temperature). Each assay included known BSE-positive and -negative control tissue sections to verify the sensitivity and specificity of the procedure. Immunohistochemical staining of rectal biopsy samples was performed as previously described, using the MoAb R145.53,56

RESULTS Confirmation of BSE infection in blood donors The seven sheep selected as blood donors for this study showed clinical signs typically associated with BSE infection, including pruritus and ataxia, with incubation periods ranging from approximately 800 to 1100 days. Brain and peripheral lymphoid tissues from each donor tested positive for the presence of abnormal prion protein by Western blotting and immunohistochemistry. Figures 1A through 1C show positive staining in the dorsal motor

EVALUATION OF P-CAPT FILTER

Fig. 1. Immunohistochemical labeling of PrPd to confirm BSE infection. This figure shows a series of representative images taken of the DMNV (brain), labeled with the PrP MoAb BG4. (A-C) From BSE-infected donor sheep N220, N245, and N218, respectively. (D-G) From recipient sheep P511, P467, P548, and P524, respectively. (H) Positive control; sheep infected with BSE. (I) Negative control; uninfected sheep. Positive PrPd labeling, comparable to that of the positive control (H), is seen in all three donors (A-C) and in the paired recipients of blood components from Donor N220, P511 (D) and P524 (G). Scale bar 5 100 mm.

nuclei of the vagus nerve (DMNV) of three donors with widespread, diffuse extracellular PrPd labeling evident in each animal. The staining pattern is consistent with that seen in the positive control section (known BSE sheep sample, Fig. 1H) and distinct from that seen in the negative control section (known uninfected control sample, Fig. 1I), where no PrPd labeling is evident. Two antibodies were used to confirm and validate the experimental outcomes (Fig. S2, available as supporting information in the online version of this paper). The same PrPd staining profile was observed in the other BSE-positive donors (not shown).

Transfusion components meet required specifications Table 2 shows the variables measured for each unit of whole blood collected from the donors and when proc-

essed to LR-RBCs and LR-RBCs-P-CAPT, including volumes and filtration times. Other variables, such as the hematocrit and percentage of plasma in different components, were measured to assess the consistency of components prepared from different donors (data not shown) and did not differ significantly from those previously reported.17 After leukoreduction, the number of WBCs in each paired unit of LR-RBCs and LR-RBCs-P-CAPT was measured by flow cytometry and by qPCR for the CD45 gene expressed by all WBCs. Table 2 shows the residual number of WBCs in the LR-RBC and LR-RBC-P-CAPT units, respectively, determined using the cytometer method. All of the units tested contained fewer than 1 3 106 WBCs/ unit and thus met the specification for human LR blood components.57 The results of the CD45 qPCR were similar for samples from all donors, with estimated numbers of WBCs approximately 3 logs less than that seen in the RBC Volume 55, September 2015 TRANSFUSION 2127

0k 0k 1.3

220

505 19

U1*

1.3 9.8

258

490 12

U2†

426‡ 15 56 286

N178

21 1.3

485 19 65 259 228

U2†

377‡ 10

U1*

N258

1.1 4.8

259

U1*

417‡ 14 52 251 503 22

U2†

50

284

4.4

502 18 45 297

14

U1*

455‡ 38 70 288

U2†

467‡ 16

U1* U2†

517 23

262

NA§

U1*

535 22 63 289

NA§

Donor ID

N180 N245 N218

Volume whole blood (mL) Leukofiltration time (min) P-CAPT filtration time (min) Volume LR-RBCs or LR-RBCsP-CAPT transfused (mL) WBC count (3104 cells/unit)

Component variables

N220

TABLE 2. Specifications of blood components transfused to recipients

* Components prepared from Unit 1 (U1) 5 LR-RBCs-P-CAPT. † Components from Unit 2 (U2) 5 LR-RBCs. ‡ Components which are lower in volume compared to the specifications for human equivalents. Acceptable volume specifications in human transfusion practice are 468 to 558 mL for a unit of whole blood and 220 to 340 mL for RBCs. § NA 5 WBC count not assessed due to equipment failure on the day these units were processed. k The 0 events recorded for the LR-RBCs from Donor N259 does not mean that the units contained no WBCs but rather that the residual numbers were below the limit of detection of the assay.

205‡

499 16

U1*

447‡ 35 40 247

U2†

N259

U2†

MCCUTCHEON ET AL.

2128 TRANSFUSION Volume 55, September 2015

TABLE 3. Results of RAMALT biopsy in transfusion recipients Recipient ID P524 P511

P548 P466

P467 P461

Q391

Q397

P494

P304

P541

P465

P473

P280

Number of positive follicles/total follicles (% positive)

Sample time (days postinfection)

29/29 (100) 2/10 (20) 11/23 (48) 2/5 (40) 0/20 0/12 0/28 0/12 0/9 NA* 0/24 0/36 0/4 0/30 0/8 0/8 0/14 0/6 0/2 0/23 0/21 0/3 0/3 0/0 0/2 0/15 0/8 0/2 0/31 0/13 0/4 0/9 0/4 0/9 0/10 0/4 0/13

676 676 763 942 587 853 587 853 1448 NA* 589 850 1445 573 839 1434 573 839 1434 530 796 1391 530 796 1391 377 643 1238 376 642 1237 313 579 1174 313 579 1174

* NA 5 Sheep P467 was euthanized before biopsies were collected.

units (data not shown), providing confirmatory evidence of adequate leukoreduction in units where it was not possible to obtain a WBC count by cytometry (Donors N220 and N259).

Infectivity associated with LR-RBCs-P-CAPT Table 3 shows the timing of and outcome of RAMALT biopsies in euthanized and surviving transfusion recipients. The only recipients found to have positive staining for abnormal PrPd in lymphoid follicles within biopsies were P524 and P511, which received paired LR-RBCs and LR-RBCs-P-CAPT components (respectively) from Donor N220. Sheep P511 had three positive biopsy results from samples collected between 676 and 942 days postinfection (representing 64%-90% of the interval between transfusion and postmortem 5 survival period), while P524 had a single positive biopsy, which was collected close to the

EVALUATION OF P-CAPT FILTER

LR RBCs from donors N245 and N218, respectively. Taken together, the results confirm that the only transfusion recipients that showed clinical and/or pathologic evidence of infection were P524 (LR-RBC recipient) and P511 (LRRBC-P-CAPT recipient), which both received components from Donor N220.

DISCUSSION

Fig. 2. Western blot detection of PrPSc for confirmation of infection in transfusion recipients. PK-resistant PrPSc was prepared from PK-digested and NAPTA-precipitated tissue homogenates. Tissues from euthanized recipients P524, P511, P467, and P548 were examined. Immunoblots were probed with the antibody ROS-BC6 (0.5 mg/mL). Tissues analyzed were medulla (diluted 1:10 before loading, Lane 1), neat extracts of spleen (Lane 2), prescapular lymph node (Lane 3), mesenteric lymph node (Lane 4), distal ileal Peyer’s patches (Lane 5), and tonsil (Lane 6). Molecular weight markers are shown in kDa on the left-hand side of the immunoblots and the exposure time was 6 minutes.

time when it developed clinical signs (92% of survival period). No further biopsies were undertaken as it has been observed that with age and in the absence of antigenic stimulus the follicles regress and become more scarce (data not shown). The same two sheep (P524 and P511) were later euthanized after the demonstration of definite or suspect clinical signs associated with BSE. Two additional recipient sheep (P548 and P467) were culled for welfare reasons. Immunohistochemistry and Western blotting performed on brain and lymphoid tissue samples confirmed that P524 and P511 were infected with BSE, while P548 and P467 were not. Figure 1 shows the pattern of abnormal prion protein deposition in the DMNV from recipient sheep P524 (Fig. 1G) and P511 (Fig. 1D), which is consistent with that seen in their donor (N220, Fig. 1A) and similar to that previously reported for sheep experimentally infected with BSE. Figure 2 shows Western blots of PKresistant PrPSc prepared from brain and lymphoid tissues from the clinically positive (P524 and P511) and clinically negative (P467 and P548) sheep. PK-resistant PrPSc was identified in brain and most lymphoid tissues tested in P524 and in brain, spleen, and ileal Peyer’s patch of P511. No PK-resistant PrPSc was detected in brain or lymphoid tissue in recipients P467 or P548, both of which received

We undertook an independent evaluation to assess the qualitative effects of prion reduction after P-CAPT filtration of LR-RBCs compared to LR-RBCs alone. Unique to this study is that the source components are endogenously infected RBCs from a large animal model. Of 14 recipients of LR-RBCs or LR-RBCs-P-CAPT, we observed infection and clinical disease in two recipients of paired components from the same donor. Owing to the limitations in blood volume that can be collected from an individual sheep at one time, we were unable to carry out transfusions of non–LR-RBCs for direct comparison of transmission rates. However, in the related leukoreduction study, similar volumes of RBCs prepared from BSE-infected donors at preclinical time points (30%-50% of incubation period) transmitted infection to eight of 21 recipients (38% attack rate), while LRRBCs transmitted to three of 12 (25%) recipients (unpublished data). There is evidence from previous sheep studies showing that the probability of transmission of BSE or scrapie by blood transfusion increases as donors progress toward the clinical phase of infection,16,58 probably as a result of increasing titers of infectivity and that RBCs collected from scrapie-infected sheep in the late preclinical period can infect 100% of recipients.41 In support of this, the donor sheep (N220) that transmitted infection in the current study also provided blood components at a preclinical time point (37% of survival period), which failed to transmit infection to their respective recipients (see Table S1, available as supporting information in the online version of this paper). If we accept that the transmission rate in recipients of non–LR-RBCs from clinical donors would be greater than 38% and possibly as high as 100%, the fact that only one of seven recipients of LR-RBCs became infected indicates that leukoreduction alone reduces the risk of transmission. Since infection was seen in the paired recipient of LR-RBCs-P-CAPT from the same donor, it may tentatively be concluded that prion reduction by filtration does not provide an additional benefit to leukoreduction in reducing the risk of transmission. However, as the numbers of animals in the study are too small to give statistical power, we are unable to provide quantitative analysis of the level of risk reduction associated with prion filtration compared to leukoreduction alone. In addition, since we lack a suitable bioassay host (e.g., transgenic mouse) for quantifying BSE infectivity in sheep blood, we cannot provide Volume 55, September 2015 TRANSFUSION 2129

MCCUTCHEON ET AL.

estimates of titer before and after leukoreduction and prion filtration that would allow calculation of prion clearance values. Our findings are similar to those reported after blood transfusions from scrapie-infected sheep,41 in which one of five recipients of LR-RBCs and one of five recipients of LR and prion-filtered RBCs showed evidence of infection when culled at the end of the experiment (but did not develop clinical disease). The latter study tested the effect of the Leukotrap (Pall) and KASEI (Asahi) filters, but not the P-CAPT filter. At present, the only peer-reviewed data available on the use of the P-CAPT filter relate to its application for removal of exogenous and endogenous infectivity associated with 263K hamster scrapie. Studies using hamster brain homogenate spiked into a full unit of human LRRBCs indicated that processing using the P-CAPT filter resulted in approximately 3-log reduction in infectivity.38 However, it has been shown that as little as 200 mL of intravenously (IV) administered blood can transmit infection between scrapie-infected sheep, despite very low infectious titers estimated by intracerebral inoculation of blood into transgenic mice expressing sheep PrP, while the equivalent minimal infectious dose of brain-derived scrapie for sheep by the IV route is approximately 1000 times greater.20 This implies that results obtained from experiments using exogenous spikes of brain-derived infectivity to measure prion clearance by various processes may not accurately predict effects on endogenous infectivity. Evaluation of the test resins used to develop the P-CAPT filter showed that a combination of leukoreduction and prion filtration reduced endogenous infectivity in a unit of hamster whole blood by more than 1.22 log ID.32 Another recent study, in which 2 units of LR-RBCs prepared from scrapie-infected hamsters were filtered using the P-CAPT device, showed an overall reduction in infectivity of approximately 1.4 log ID, but concluded that there was residual infectivity of around 0.2 ID/mL in the prionfiltered blood (Dr J.M. Sutton, written communication, September 2014). Leukoreduction of scrapie-infected hamster blood alone has been estimated to remove up to approximately 70% of endogenous infectivity, with the remainder being plasma associated. However, transfusion experiments using prion-infected sheep and deer suggest that the infectivity in plasma is much lower in these species and may be largely cell associated.17,19,41 The single case of “subclinical” vCJD infection identified in a person with hemophilia was thought most likely to have been caused by documented exposure to plasma products derived from a vCJD patient, which suggests that human plasma can also be infectious.28 These apparent species differences may reflect host-specific variation in the distribution of endogenous infectivity or may arise from differences in the protocols used for preparation of components in the various studies. In this study, it is possible that the trans2130 TRANSFUSION Volume 55, September 2015

mission of infection by LR-RBC and LR-RBC-P-CAPT components could be explained by infectivity associated with residual plasma and/or the small numbers of WBCs remaining after leukoreduction and prion filtration. However, comparable levels of leukoreduction were achieved in components from all seven donors, but only one (N220) transmitted the infection, suggesting that this sheep may have had unusually high titers of blood-borne infectivity. Ten of the sheep that were transfused with either LRRBCs or LR-RBCs-P-CAPT are alive, with current survival periods ranging from 1710 to 1984 days postinfection (Table 1). There is as yet no clear evidence, from clinical observations and the results of rectal biopsies, that any of these sheep are infected with BSE. The incubation periods of confirmed BSE-infected transfusion recipients from our ongoing leukoreduction study17 were 1018 6 188 days (mean 6 SD) for 141LF genotype sheep (n 5 25) and 705 6 120 days for 141FF sheep (n 5 15), respectively (unpublished data). For comparison, the majority (9/10) of the surviving recipients in this experiment are 141LF (mean 6 SD survival period, 1884 6 118 days postinfection), with a single 141FF recipient (survival period, 1773 days postinfection). These survival periods represent a difference from the observed incubation periods of more than 4 SDs of the mean, suggesting that it is unlikely that the remaining sheep will develop clinical signs. However, it is possible that they will show signs of infection when they are culled, even in the absence of clinical disease. Their prolonged survival after transfusion is consistent with a substantial reduction in infectivity associated with leukoreduction alone or in combination with prion filtration. In conclusion, we have demonstrated that, for one out of seven BSE-infected sheep used as blood donors, PCAPT filtration of LR-RBCs did not remove all endogenous infectivity, resulting in infection and clinical disease in the transfused recipient. This supports the view that the filter cannot be assumed to prevent transmission of vCJD by blood transfusion. ACKNOWLEDGMENTS The authors are grateful to Hugh Simmons and colleagues (Animal Plant and Health Agency [APHA], formerly AHVLA) for the provision of sheep; the staff at the animal facilities at both The Roslin Institute and Institute for Animal Health for scientific assistance and care with all animals used in this study; Paula Stewart for genotype analysis; and Christopher de Wolf and Allister Smith for technical assistance with sample preparation and Western blotting. We also thank Maurice Bardsley and the Biological Archive Group, APHA (formerly AHVLA) for providing the BSE-infected cattle brain homogenate used to infect donors. Significant input into the study design was provided by former colleagues at the Scottish National Blood Transfusion Service

EVALUATION OF P-CAPT FILTER

(SNBTS), specifically Dr Christopher V. Prowse, Dr Valerie Horn-

zas CI, Fournier JG, Nouvel V, et al. Adaptation of 13. Lasme

sey, and Dr Ian McGregor and we especially acknowledge their

the bovine spongiform encephalopathy agent to primates

valuable contribution. SMC, JM, MLT, and other representatives from the SNBTS conceived and designed the experiments; SMC,

and comparison with Creutzfeldt-Jakob disease: implications for human health. Proc Natl Acad Sci U S A 2001;98:

EFH, ARAB, BCT, AS, LG, SM, GM, and NA performed the experi-

4142-7.

ments; SMC and ARAB analyzed the data; SMC wrote the publication; and all authors contributed to editing and approval of the

14. Houston F, Foster JD, Chong A, et al. Transmission of BSE by blood transfusion in sheep. Lancet 2000;356:999-1000.

final version of the report.

15. Hunter N, Foster J, Chong A, et al. Transmission of prion

CONFLICT OF INTEREST The authors have disclosed no conflicts of interest.

REFERENCES 1. Will RG, Ironside JW, Zeidler M, et al. A new variant of Creutzfeldt-Jakob disease in the UK. Lancet 1996;347:921-5. 2. Ironside JW, Sutherland K, Bell JE, et al. A new variant of Creutzfeldt-Jakob disease: neuropathological and clinical features. Cold Spring Harb Symp Quant Biol 1996;61:523-30. 3. Bruce ME, Will RG, Ironside JW, et al. Transmissions to mice indicate that ‘new variant’ CJD is caused by the BSE agent. Nature 1997;389:498-501. 4. Diack AB, Head MW, McCutcheon S, et al. Variant CJD. 18 years of research and surveillance. Prion 2014;8:286-95. 5. Creutzfeldt-Jakob disease in the UK (by calendar year) [Internet]. Edinburgh: The National CJD Research & Surveillance Unit (NCJDRSU); updated 2014 Jul 4 [cited 2014 Nov 3]. Available from: http://www.cjd.ed.ac.uk/documents/figs. pdf 6. Brown P, Rohwer RG, Dunstan BC, et al. The distribution of infectivity in blood components and plasma derivatives in experimental models of transmissible spongiform encephalopathy. Transfusion 1998;38:810-6. kova L, McShane LM, et al. Further studies 7. Brown P, Cervena of blood infectivity in an experimental model of transmissible spongiform encephalopathy, with an explanation of why blood components do not transmit Creutzfeldt-Jakob disease in humans. Transfusion 1999;39:1169-78. 8. Cervenakova L, Yakovleva O, McKenzie C, et al. Similar levels of infectivity in the blood of mice infected with humanderived vCJD and GSS strains of transmissible spongiform encephalopathy. Transfusion 2003;43:1687-94. 9. Bons N, Lehmann S, Mestre-Frances N, et al. Brain and buffy coat transmission of bovine spongiform encephalopathy to the primate Microcebus murinus. Transfusion 2002;42:513-6. 10. Herzog C, Sale`s N, Etchegaray N, et al. Tissue distribution of bovine spongiform encephalopathy agent in primates after intravenous or oral infection. Lancet 2004;363:422-8. 11. Holada K, Vostal JG, Theisen PW, et al. Scrapie infectivity in hamster blood is not associated with platelets. J Virol 2002; 76:4649-50. 12. Taylor DM, Fernie K, Reichl HE, et al. Infectivity in the blood of mice with a BSE-derived agent. J Hosp Infect 2000;46: 78-9.

diseases by blood transfusion. J Gen Virol 2002;83: 2897-905. 16. Houston F, McCutcheon S, Goldmann W, et al. Prion diseases are efficiently transmitted by blood transfusion in sheep. Blood 2008;112:4739-45. 17. McCutcheon S, Blanco ARA, Houston EF, et al. All clinicallyrelevant blood components transmit prion disease following a single blood transfusion: a sheep model of vCJD. PLoS One 2011;6:e23169. 18. Mathiason CK, Powers JG, Dahmes SJ, et al. Infectious prions in the saliva and blood of deer with chronic wasting disease. Science 2006;314:133-6. 19. Mathiason CK, Hayes-Klug J, Hays SA, et al. B cells and platelets harbor prion infectivity in the blood of deer infected with chronic wasting disease. J Virol 2010;84: 5097-107. oletti O, Litaise C, Simmons H, et al. Highly efficient 20. Andre prion transmission by blood transfusion. PLoS Pathog 2012; 8:e1002782. 21. Hewitt PE, Llewelyn CA, Mackenzie J, et al. Creutzfeldt-Jakob disease and blood transfusion: results of the UK Transfusion Medicine Epidemiological Review study. Vox Sang 2006;91: 221-30. 22. Llewelyn CA, Hewitt PE, Knight RS, et al. Possible transmission of variant Creutzfeldt-Jakob disease by blood transfusion. Lancet 2004;363:417-21. 23. Wroe SJ, Pal S, Siddique D, et al. Clinical presentation and pre-mortem diagnosis of variant Creutzfeldt-Jakob disease associated with blood transfusion: a case report. Lancet 2006;368:2061-7. 24. Health protection report: news archives (Vol 1, No 3, 19 January 2007). Fourth case of transfusion-associated variantCJD infection [Internet]. London: Public Health England; 2007 [cited 2014 Dec 16]. Available from: http://webarchive. nationalarchives.gov.uk/20131102034040/http://www.hpa. org.uk/hpr/archives/2007/news2007/news0307.htm 25. Hewitt PE, Llewelyn CA, Mackenzie J, et al. Three reported cases of variant Creutzfeldt-Jakob disease transmission following transfusion of labile blood components. Vox Sang 2006;91:348. 26. Ironside JW. Variant Creutzfeldt-Jakob disease: an update. Folia Neuropathol 2012;50:50-6. 27. Peden AH, Head MW, Ritchie DL, et al. Preclinical vCJD after blood transfusion in a PRNP codon 129 heterozygous patient. Lancet 2004;364:527-9. 28. Peden A, McCardle L, Head MW, et al. Variant CJD infection in the spleen of a neurologically asymptomatic Volume 55, September 2015 TRANSFUSION 2131

MCCUTCHEON ET AL.

UK adult patient with haemophilia. Haemophilia 2010;16: 296-304. 29. Bishop MT, Hart P, Aitchison L, et al. Predicting susceptibility and incubation time of human-to-human transmission of vCJD. Lancet Neurol 2006;5:393-8. 30. Gill ON, Spencer Y, Richard-Loendt A, et al. Prevalent abnormal prion protein in human appendixes after bovine spongiform encephalopathy epizootic: large scale survey. BMJ 2013;347:f5675. 31. VCJD: further precautionary measures announced [Internet]. London: Public Health England; 2004 21 Sep [cited 2014 Dec 16]. Available from: http://webarchive.nationalarchives.gov. uk/1/www.dh.gov.uk/en/Publicationsandstatistics/Pressreleases/DH_4089689 32. Gregori L, Gurgel PV, Lathrop JT, et al. Reduction in infectivity of endogenous transmissible spongiform encephalopathies present in blood by adsorption to selective affinity resins. Lancet 2006;368:2226-30. 33. Gregori L, McCombie N, Palmer D. Effectiveness of leucoreduction for removal of infectivity of transmissible spongiform encephalopathies from blood. Lancet 2004;364: 529-31. 34. Tateishi J, Kitamoto T, Mohri S, et al. Scrapie removal using Planova virus removal filters. Biologicals 2001;29:1725. 35. Yunoki M, Tanaka H, Urayama T, et al. Infectious prion protein in the filtrate even after 15 nm filtration. Biologicals 2010;38:311-3. 36. Poelsler G, Berting A, Kindermann J, et al. A new liquid intravenous immunoglobulin with three dedicated virus reduction steps: virus and prion reduction capacity. Vox Sang 2008;94:184-92. 37. Roberts PL, Dalton J, Evans D, et al. Removal of TSE agent from plasma products manufactured in the United Kingdom. Vox Sang 2013;104:299-308. 38. Lescoutra-Etchegaray N, Sumian C, Culeux A, et al. Removal of exogenous prion infectivity in leukoreduced red blood cells unit by a specific filter designed for human transfusion. Transfusion 2014;54:1037-45. 39. Gregori L, Lambert BC, Gurgel PV, et al. Reduction of transmissible spongiform encephalopathy infectivity from human red blood cells with prion protein affinity ligands. Transfusion 2006;46:1152-61. 40. Cardone F, Sowemimo-Coker S, Abdel-Haq H, et al. Assessment of prion reduction filters in decreasing infectivity of ultracentrifuged 263K scrapie-infected brain homogenates in “spiked” human blood and red blood cells. Transfusion 2014;54:990-5. 41. Lacroux C, Bougard D, Litaise C, et al. Impact of leucocyte depletion and prion reduction filters on TSE blood borne

cell culture-based infectivity assay. Transfusion 2010;50:9808. 43. Elebute MO, Choo L, Mora A, et al. Transfusion of prionfiltered red cells does not increase the rate of alloimmunization or transfusion reactions in patients: results of the UK trial of prion-filtered versus standard red cells in surgical patients (PRISM A). Br J Haematol 2013;160:701-8. 44. Cancelas JA, Rugg N, Pratt PG, et al. Infusion of P-Capt prion-filtered red blood cell products demonstrate acceptable in vivo viability and no evidence of neoantigen formation. Transfusion 2011;51:2228-36. 45. Hornsey VS, Casey C, McColl K, et al. Characteristics of prion-filtered red cells suspended in pathogen-inactivated plasma (MB treated or solvent-detergent treated) for neonatal exchange transfusion. Vox Sang 2011;101:28-34. 46. Murphy CV, Eakins E, Fagan J, et al. In vitro assessment of red-cell concentrates in SAG-M filtered through the MacoPharma P-CAPT prion-reduction filter. Transfus Med 2009; 19:109-16. 47. Wiltshire M, Thomas S, Scott J, et al. Prion reduction of red blood cells: impact on component quality. Transfusion 2010; 50:970-9. 48. Peden AH, Ironside JW. Review: pathology of variant Creutzfeldt-Jakob disease. Folia Neuropathol 2004;42 Suppl A:85-91. 49. Jeffrey M, Ryder S, Martin S, et al. Oral inoculation of sheep with the agent of bovine spongiform encephalopathy (BSE). 1. Onset and distribution of disease-specific PrP accumulation in brain and viscera. J Comp Pathol 2001;124:280-9. 50. Simmons HA, Simmons MM, Spencer YI, et al. Atypical scrapie in sheep from a UK research flock which is free from classical scrapie. BMC Vet Res 2009;5:8. 51. Tan BC, Blanco AR, Houston EF, et al. Significant differences in incubation times in sheep infected with bovine spongiform encephalopathy result from variation at codon 141 in the PRNP gene. J Gen Virol 2012;93:2749-56. lez L, Jeffrey M, Dagleish MP, et al. Susceptibility to 52. Gonza scrapie and disease phenotype in sheep: cross-PRNP genotype experimental transmissions with natural sources. Vet Res 2012;43:55. lez L, Dagleish MP, Martin S, et al. Diagnosis of pre53. Gonza clinical scrapie in live sheep by the immunohistochemical examination of rectal biopsies. Vet Rec 2008;162:397-403. 54. McCutcheon S, Hunter N, Houston F. Use of a new immunoassay to measure PrP Sc levels in scrapie-infected sheep brains reveals PrP genotype-specific differences. J Immunol Methods 2005;298:119-28. 55. McCutcheon S, Langeveld JP, Tan BC, et al. Prion proteinspecific antibodies that detect multiple TSE agents with high

transmission. PLoS One 2012;7:e42019. 42. Sowemimo-Coker SO, Demczyk CA, Andrade F, et al. Evalua-

sensitivity. PLoS One 2014;9:e91143. lez L, Martin S, Houston FE, et al. Phenotype of 56. Gonza disease-associated PrP accumulation in the brain of bovine

tion of removal of prion infectivity from red blood cells with

spongiform encephalopathy experimentally infected sheep.

prion reduction filters using a new rapid and highly sensitive

J Gen Virol 2005;86:827-38.

2132 TRANSFUSION Volume 55, September 2015

EVALUATION OF P-CAPT FILTER

57. JPAC—Joint United Kingdom (UK) Blood Transfusion and Tissue Transplantation Services Professional Advisory Committee. Guidelines for the blood transfusion services in the UK [Internet]. 8th ed. London: NHS Blood and Transplant; 2013 [cited 2014 May 28]. Available from: http://www.transfusionguidelines.org.uk/red-book ndez-Borges N, et al. Prionemia 58. Lacroux C, Vilette D, Ferna and leukocyte-platelet-associated infectivity in sheep transmissible spongiform encephalopathy models. J Virol 2012; 86:2056-66.

SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article at the publisher’s website: Fig. S1. Illustrative overview of blood processing: leukoreduction and P-CAPT filtration. This scheme illustrates the process by which units of whole blood were processed to leukoreduced red cell concentrates and PCAPT filtered equivalents that were used for transfusion to recipients. Two units of whole blood, in anticoagulant, were collected from each donor sheep, centrifuged and red cell concentrates (RCC) extracted. Each unit of RCC was independently passed through leukoreduction (LR) filters and then units pooled to ensure homogeneity in terms of cellular content and distribution of BSE-

associated blood-borne infectivity. The pooled volume was split, whereby one unit of LR-RCC was used for transfusion and the other LR-RCC processed through the P-CAPT filter prior to transfusion. Fig. S2. Confirmation of BSE infection in P-CAPT donors and recipients by staining hypoglossal nuclei with ROS-IH9. These images are representative images of formalin-fixed tissue sections of hypoglossal nuclei (brain) from blood donors and recipients stained with the anti-prion protein antibody ROS-IH9. Panels A and C are the donor and LR-RBC recipient pair N220 and P524 respectively, and show an intense, punctate (sometimes coalescing in nature) PrPd labelling profile (brown staining) within the cell body of the neurones as well as some punctate staining out with the cell body. Both staining profiles are indicative of BSE infection. Panels B and C represent the donor and LR-RBC recipient pair N245 and P467 respectively. The donor (panel B) was confirmed as having BSE, indicated by the PrPd labelling pattern previously described, whereas the recipient (panel D) showed no specific PrPd labelling, characteristic of this antibody, as was confirmed as negative (alongside the results from other tests) for BSE infection. Scale bar 5 100 lm. Table S1. Correlation of donors ID used this study and coded IDs published previously

Volume 55, September 2015 TRANSFUSION 2133